CN111582536A - Hidden fault prediction method, device, equipment and medium based on feature learning - Google Patents
Hidden fault prediction method, device, equipment and medium based on feature learning Download PDFInfo
- Publication number
- CN111582536A CN111582536A CN202010206984.4A CN202010206984A CN111582536A CN 111582536 A CN111582536 A CN 111582536A CN 202010206984 A CN202010206984 A CN 202010206984A CN 111582536 A CN111582536 A CN 111582536A
- Authority
- CN
- China
- Prior art keywords
- fault
- trend
- clustering
- zone
- extension
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 63
- 238000005065 mining Methods 0.000 claims abstract description 77
- 239000006185 dispersion Substances 0.000 claims abstract description 32
- 238000011161 development Methods 0.000 claims abstract description 23
- 239000003245 coal Substances 0.000 claims abstract description 22
- 238000010276 construction Methods 0.000 claims description 24
- 238000004519 manufacturing process Methods 0.000 claims description 24
- 238000004458 analytical method Methods 0.000 claims description 19
- 238000012545 processing Methods 0.000 claims description 19
- 238000012417 linear regression Methods 0.000 claims description 10
- 238000003860 storage Methods 0.000 claims description 9
- 238000012512 characterization method Methods 0.000 claims description 7
- 238000010606 normalization Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 3
- 238000003672 processing method Methods 0.000 claims description 3
- 238000009412 basement excavation Methods 0.000 abstract description 6
- 238000010586 diagram Methods 0.000 description 20
- 238000004364 calculation method Methods 0.000 description 15
- 238000004891 communication Methods 0.000 description 11
- 230000008569 process Effects 0.000 description 10
- 238000007621 cluster analysis Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 238000009826 distribution Methods 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 4
- 238000000605 extraction Methods 0.000 description 3
- 210000000887 face Anatomy 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 210000000349 chromosome Anatomy 0.000 description 1
- 238000013016 damping Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000005553 drilling Methods 0.000 description 1
- 238000005194 fractionation Methods 0.000 description 1
- 230000001939 inductive effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000011158 quantitative evaluation Methods 0.000 description 1
- 210000003786 sclera Anatomy 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/02—Agriculture; Fishing; Forestry; Mining
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Software Systems (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Evolutionary Computation (AREA)
- General Business, Economics & Management (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Operations Research (AREA)
- Entrepreneurship & Innovation (AREA)
- Mining & Mineral Resources (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Animal Husbandry (AREA)
- Agronomy & Crop Science (AREA)
- Quality & Reliability (AREA)
- Marine Sciences & Fisheries (AREA)
- Medical Informatics (AREA)
- Game Theory and Decision Science (AREA)
- Computing Systems (AREA)
- Development Economics (AREA)
- Mathematical Physics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
One or more embodiments of the present specification provide a method, an apparatus, a device, and a medium for predicting hidden faults based on feature learning, where the method includes: fault attribute information is obtained according to the exposure information of the mining area, and the fault attribute information is processed according to a preset rule to obtain fault sample data; clustering the fault sample data by using trend characteristics to obtain each dominant trend cluster of the well field fault development, wherein the optimal grouping number of the trend characteristics is determined by a contour coefficient; and on the basis of fault strike characteristic clustering, clustering each dominant strike cluster again by taking the distance defined for the fault as a characteristic to obtain each fault zone developed in the well field. The invention identifies the fault zone by adopting the trend and distance characteristics, defines the extension index, the buffer radius and the fall expectation of the fault zone to respectively depict the extension, the dispersion in the trend and the fall characteristics of the fault zone in the trend, and realizes the quantitative prediction of the hidden fault in front of the coal mine excavation working face.
Description
Technical Field
One or more embodiments of the present disclosure relate to the field of geological exploration technologies, and in particular, to a hidden fault prediction method, device, apparatus, and medium based on feature learning.
Background
Faults are the most common geological structures in coal mine production, not only cause coal resource waste, but also are direct reasons for inducing roof accidents, gas outbursts and water damage accidents. Currently, large faults with a throw greater than 20 meters have been generally ascertained during the exploration phase, and they are generally used to demarcate a field or production zone. However, medium and small scale faults have not been accurately identified under the existing exploration technical conditions, and they are mainly discovered through production disclosure. Therefore, the prediction of medium and small scale blind faults, especially faults with a fall of less than 10 meters, is the focus of coal mine geological work.
In order to find out the hidden fault in front of the mining working face, an effective means is to adopt various means such as geophysical prospecting and drilling to carry out advanced detection. However, the advanced detection is limited by the construction cost and construction conditions of the engineering, and is pointed instead of being combined with the development of production from time to place. The fault which can be hidden in front of the mining working face is predicted by analyzing the development rule of the revealed fault, so that the problems are solved. Scholars at home and abroad research the number, scale and spatial distribution characteristics of faults and the correlation between spatial distribution indexes and fault scale indexes by adopting a fractal theory. The random fault simulation method is also used by scholars to draw a fault existence probability distribution map. In addition, the artificial neural network is introduced into fault development rule analysis and prediction, the abnormal indexes of the coal bed floor elevation, the coal thickness, the water inflow amount, the gas emission amount and the like related to fault development are used as the basis for predicting the existence of faults, data disclosed by a mining area is used as a training sample, a prediction model is obtained through machine learning, and finally, the judgment of whether the faults exist in front of a mining working face can be given.
The scientific research results increase the structural complexity of the coal mine excavation area from qualitative recognition to quantitative evaluation, and also increase the evaluation from the quantitative complexity evaluation to the prediction of whether hidden faults exist in front of excavation, but the positions and the attributes of the faults existing in front of excavation cannot be predicted, so that the application values of the faults are limited undoubtedly.
Disclosure of Invention
In view of the above, one or more embodiments of the present disclosure are directed to a hidden fault prediction method, apparatus, device, and medium based on feature learning, in which fault zones are used as vectors for expressing fault development rules, relevant attributes of the fault zones are defined, and probabilities and location deviation ranges of predictions and hidden faults are quantitatively given, so as to solve problems existing in current hidden fault predictions.
In view of the above, in a first aspect, one or more embodiments of the present specification provide a hidden fault prediction method based on feature learning, the method including:
acquiring fault attribute information according to the exposure information of the mining area, wherein the fault attribute information at least comprises fault position, fall, trend and extension length attribute information, and processing the fault attribute information according to a preset rule to acquire fault sample data;
clustering the fault sample data by using trend characteristics to obtain each dominant trend cluster of the well field fault development, wherein the optimal grouping number of the trend characteristics is determined by a contour coefficient;
clustering each dominant trend cluster again by taking the distance defined for the fault as a characteristic on the basis of fault trend characteristic clustering to obtain each fault zone developed in a well field, wherein the optimal grouping number of the fault distance characteristic is determined by a contour coefficient;
and obtaining the central line of each fault zone by adopting a linear regression method, calculating the extension index, the buffer radius and the expected fall attribute value of each fault zone to characterize the extension, the dispersion of the trend and the fall of the fault zone, and predicting the hidden fault in front of the mining working face.
In another possible implementation manner of the example of the present invention, in combination with the above description, before the calculating the elongation index, the buffer radius, and the drop height expected attribute value of each fault zone, the method further includes:
defining the extension index, the buffer radius and the drop height expectation of each fault zone respectively, wherein the definition comprises the following steps:
defining the extension index, wherein the extension index is the exposure proportion of the fault zone in the mined range;
defining the extension index, wherein the extension index is the swing radius of the fault belt in the fault belt structure middle line; and
defining the throw expectation, which is an expected value of a fault throw included in the fault zone.
In another possible implementation manner of the example of the present invention, before clustering again each dominant trend cluster to characterize the distance defined by the fault, the method further includes:
determining the datum line according to the determined mining area and the trend value of the fault zone of the mining area, wherein the datum line comprises the following steps:
when the trend value of the construction line is in a first preset range, determining the datum line by combining the space coordinate of one end point of the mining area and the trend average value of all faults of each set of fault sample data;
and when the trend value of the construction line is in a second preset range, determining the datum line by combining the spatial coordinate of the other end point of the mining area and the trend average value of all faults of each set of fault sample data.
In another possible implementation manner of the example of the present invention, the obtaining the centerline position of each fault zone by linear regression and calculating each attribute value defined by the fault zone includes:
determining the dispersion of sample data of each fault included in each fault zone, wherein the dispersion is the ratio of the sample standard deviation of the distance of the fault offset construction central line to the projection length of all faults included in the fault zone on the construction central line;
and predicting whether the fault precision meets the requirement of guiding production according to the relation between the dispersion of the fault zone classification result and a preset value.
With reference to the above description, in another possible implementation manner of the example of the present invention, the processing the fault data according to the preset rule to obtain fault sample data includes:
taking the ratio of fault fall to the average coal thickness of the mining working face and the extension length of the fault as characterization indexes of the fault data;
processing the characterization indexes of the fault data according to a piecewise linear normalization processing method based on a first preset threshold and a second preset threshold to obtain fault development scale weight;
and clustering the fault data by taking the fault development scale weight as a clustering feature, and taking the fault which has a certain influence on the production in a clustering result as fault sample data of next-step clustering analysis.
In a second aspect, the present invention also provides a hidden fault prediction apparatus based on feature learning, the apparatus comprising:
the data processing module is used for obtaining fault attribute information according to the exposure information of the mining area, wherein the fault attribute information at least comprises fault position, fall, trend and extension length attribute information, and processing the fault attribute information according to a preset rule to obtain fault sample data;
the first clustering module is used for clustering fault sample data according to trend characteristics to obtain each dominant trend cluster of the well field fault development, and the optimal grouping number of the trend characteristics is determined by a contour coefficient;
the second clustering module is used for clustering each dominant trend cluster again by taking the distance defined for the fault as a characteristic on the basis of fault trend characteristic clustering to obtain each fault zone developed in the well field, wherein the optimal grouping number of the fault distance characteristic is determined by a contour coefficient;
and the analysis and prediction module is used for obtaining the central line of each fault zone by adopting a linear regression method for each fault zone obtained by the second clustering module, calculating the extension index, the buffer radius and the expected fall attribute value of each fault zone, describing the extension, the dispersion of the trend and the fall characteristic of the fault zone, and predicting the hidden fault in front of the mining working face.
The above apparatus, further comprising: a definition module to:
defining the extension index, the buffer radius and the drop height expectation of each fault zone respectively, wherein the definition comprises the following steps:
defining the extension index, wherein the extension index is the exposure proportion of the fault zone in the mined range;
defining the extension index, wherein the extension index is the swing radius of the fault belt in the fault belt structure middle line; and
defining the throw expectation, which is an expected value of a fault throw included in the fault zone.
The above apparatus, further comprising:
the datum line determining module is used for determining the datum line according to the determined mining area and the trend value of the fault zone of the mining area; the datum line determining module comprises:
the first sub-module is used for determining the datum line by combining the space coordinate of one end point of the mining area and the trend average value of all faults of each set of fault sample data when the trend value of the construction line is in a first preset range;
and the second sub-module is used for determining the datum line by combining the space coordinate of the other end point of the mining area and the trend average value of all faults of each group of fault sample data when the trend value of the construction line is in a second preset range.
In a third aspect, an embodiment of the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor executes the computer program to implement the above hidden fault prediction method based on feature learning.
In a fourth aspect, embodiments of the present invention also provide a non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the above-described feature learning-based hidden fault prediction method.
As can be seen from the above description, the hidden fault prediction method, device, apparatus, and medium based on feature learning provided in one or more embodiments of the present disclosure can divide a fault into three levels according to the influence degree of the fault on mining, use the fault having a certain influence on production as valuable fault sample data, identify a fault zone developing in a mining area in a subsequent analysis process by using a feature learning machine learning method, and simultaneously calculate an extension index, a buffer radius, and a fall expectation of the fault zone, accurately characterize the extension of the fault zone, the dispersion of the trend, the fall, and the like, and provide a quantitative prediction result of the hidden fault in front of a mining working face.
Drawings
In order to more clearly illustrate one or more embodiments or prior art solutions of the present specification, the drawings that are needed in the description of the embodiments or prior art will be briefly described below, and it is obvious that the drawings in the following description are only one or more embodiments of the present specification, and that other drawings may be obtained by those skilled in the art without inventive effort from these drawings.
FIG. 1 is a schematic flow diagram of a method for feature learning based hidden fault prediction in accordance with one or more embodiments of the present disclosure;
FIG. 2 is a schematic flow chart illustrating the determination of a reference line in a hidden fault prediction method based on feature learning according to one or more embodiments of the present disclosure;
FIG. 3 is a schematic flow diagram of a method for feature learning based hidden fault prediction in accordance with one or more embodiments of the present disclosure;
FIG. 4 is a schematic diagram illustrating baseline extraction for a latent fault prediction method based on feature learning according to one or more embodiments of the present disclosure;
FIG. 5 is a schematic view of fault prediction in accordance with one or more embodiments of the present disclosure;
FIG. 6 is a schematic diagram of a field excavation according to one or more embodiments of the present disclosure;
FIG. 7 is a graph illustrating contour coefficient results for trend feature clustering in accordance with one or more embodiments of the present disclosure;
FIG. 8 is a schematic diagram of trend clustering results in one or more embodiments of the present disclosure;
FIG. 9(a) is a schematic diagram of the clustering profile coefficient results of the trend cluster 1 distance feature in one or more embodiments of the present disclosure;
FIG. 9(b) is a schematic diagram of the cluster trend 2 distance feature clustering profile coefficient results according to one or more embodiments of the present disclosure;
FIG. 9(c) is a schematic diagram of the cluster trend 3 distance feature clustering profile coefficient results according to one or more embodiments of the present disclosure;
FIG. 10(a) is a diagram illustrating a distance feature clustering result 1 according to one or more embodiments of the present disclosure;
FIG. 10(b) is a diagram illustrating a distance feature clustering result 2 according to one or more embodiments of the present disclosure;
FIG. 10(c) is a diagram illustrating a distance feature clustering result 2 according to one or more embodiments of the present disclosure;
FIG. 11 is a schematic illustration of blind fault prediction according to one or more embodiments of the present disclosure;
FIG. 12 is a schematic diagram of a latent fault prediction device based on feature learning according to one or more embodiments of the present disclosure;
fig. 13 is a schematic structural diagram of an electronic device according to one or more embodiments of the present disclosure.
Detailed Description
For the purpose of promoting a better understanding of the objects, aspects and advantages of the present disclosure, reference is made to the following detailed description taken in conjunction with the accompanying drawings.
It is to be noted that unless otherwise defined, technical or scientific terms used in one or more embodiments of the present specification should have the ordinary meaning as understood by those of ordinary skill in the art to which this disclosure belongs. The use of "first," "second," and similar terms in one or more embodiments of the specification is not intended to indicate any order, quantity, or importance, but rather is used to distinguish one element from another. The word "comprising" or "comprises", and the like, means that the element or item listed before the word covers the element or item listed after the word and its equivalents, but does not exclude other elements or items. The terms "connected" or "coupled" and the like are not restricted to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "upper", "lower", "left", "right", and the like are used merely to indicate relative positional relationships, and when the absolute position of the object being described is changed, the relative positional relationships may also be changed accordingly.
The invention relates to a hidden fault prediction method, a device, equipment and a medium based on feature learning, which are mainly applied to a scene of exploring a geological structure of a mining area in mining production, and the basic idea is as follows: the fault is divided into three levels according to the influence degree of the fault on mining, the fault which has certain influence on production is used as valuable fault sample data, multiple clustering is carried out in the subsequent analysis process according to the trend characteristics and the distance characteristics, fault zones are obtained after secondary clustering, the relevant attributes of the fault zones are defined and calculated, the fault zones are extended from the mined area to the non-mined area, the prediction of the hidden fault is realized, and the dynamic prediction of the fault in front of the mining working face is also realized by combining with the production dynamic identification of the fault zones.
The blind fault in the embodiment of the invention is a fault developed in a coal seam.
The present embodiment is applicable to a case where the intelligent terminal of the cluster analysis module performs hidden fault prediction based on feature learning, where the method may be executed by a cluster analysis apparatus, where the apparatus may be implemented by software and/or hardware, and may be generally integrated in an intelligent terminal, or may be controlled by a central control module in the terminal, as shown in fig. 1, which is a basic flow diagram of the hidden fault prediction method based on feature learning of the present invention, and the method specifically includes the following steps:
in step 110, fault attribute information is obtained according to the disclosure information of the mining area, wherein the fault attribute information at least comprises fault position, fall, trend and extension length attribute information, and the fault attribute information is processed according to a preset rule to obtain fault sample data;
the disclosure information is related geological data in a mining area acquired by a geological means before mining, and related mining data determined by a certain technical means in the process of mining, and may include average coal seam thickness, mining working face, fault number, fault fall, fault strike, fault position coordinates (two-dimensional coordinates or three-dimensional coordinates according to calculation requirements), and the like of the mining area.
According to the need, the relevant fault data is determined from the exposure information, and in an implementation manner of the embodiment of the present invention, the fault data may be the number of fault strips, fault fall, fault strike, fault position coordinates necessary for performing cluster analysis by using the distance feature value, and the like.
In a feasible implementation manner of the embodiment of the present invention, the fault attribute information at least includes attribute information of a fault, such as a position, a fall, a trend, and an extension length.
And processing the fault data according to a preset rule to obtain fault sample data, wherein the preset rule can be a process of screening the determined fault data, the determined fault data comprises fault throw and coal seam thickness, the fault throw and coal seam thickness are taken as one of indexes representing the influence degree of the fault on production, and the fault extension length is taken as a second representation index.
In a practical implementation manner of the embodiment of the invention, for coal mining, the relationship between the fault drop and the coal seam thickness directly determines the influence degree of the fault on production, so that the ratio of the fault drop to the coal seam thickness can be used as an index for representing the influence degree of the fault on the production.
Processing the characterization index of the fault data according to a piecewise linear normalization processing method based on a first preset threshold and a second preset threshold to obtain a weight of the fault development scale;
and dividing the influence of the fault on the production into 3 grades by taking 0.5 as the first preset threshold and 0.2 as the second preset threshold. The fault above the first preset threshold, namely greater than 0.5, has a larger influence, a certain influence exists between the first preset threshold 0.5 and the second preset threshold 0.2, and the corresponding level is 2 levels with larger influence (greater than 0.5) from high to low; has certain influence (between 0.2 and 0.5), and is grade 1; the production is not influenced basically (less than 0.2), and the grade is 0.
And the extension length of the fault is the length of the fault in the direction of the trend, and a second weight of the fault development scale is obtained. And clustering the fault data by taking the sum of the weights of the two fault development scales as a clustering characteristic, and taking the fault which has a certain influence on the production in a clustering result as fault sample data of next-step clustering analysis.
In an implementation manner of the embodiment of the present invention, a piecewise linear normalization method may be adopted to process a first fault development scale weight, and a calculation formula is as follows:
in the formula (1), WiIs the weight of the ith fault sample data, TiIs the fall of the ith fault sample data, n is the number of the fault sample data, Th is the average coal thickness of the mining area, TminIs the minimum fall of the fault sample, TmaxIs the maximum fall of the fault sample.
When faults are classified, the feature of fault extension length is used as a weight II, after the feature is merged with fall attribute weight to calculate the weight value of each fault, fault samples are clustered into 3 grades according to the weight feature, wherein 2 grades are faults with larger influence, 1 grade is faults with certain influence, and 0 grade is faults with smaller influence.
In the embodiment of the invention, the fault data included in the clustering result is taken as fault sample data, and 1 and 2 levels of fault samples are adopted in the next step of fault zone characteristic identification.
In step 120, clustering the fault sample data by using trend characteristics to obtain each dominant trend cluster of the well fault development, wherein the optimal grouping number of the trend characteristics is determined by a contour coefficient;
in the implementation mode of the embodiment of the invention, the clustering analysis result of the fault trend characteristic is evaluated by taking the profile coefficient as an index, the profile coefficient is used for representing the clustering degree of the clustering result, and the higher the profile coefficient is, the more the clustering is consistent with the reality.
In order to determine the most appropriate trend grouping number by using the contour coefficient as an evaluation index, the fault sample data is subjected to the cluster analysis of the trend characteristics, and in a feasible implementation manner of the embodiment of the invention, the fault sample is subjected to the cluster analysis of the trend characteristics by using a K-MEANS algorithm of a partitioning method. Different from conventional distance calculation, the fault trend in the embodiment of the invention is expressed by periodic variables, the value taking domain is [0, pi ], the fault trend is converted into an angle [0, 180 degrees ], and the value 0 is the same as pi to complete a period, which is suitable for specific mining work.
The distance calculation formula used is as follows:
d＝min(|x1-x2|,π-|x1-x2|) (2)
wherein x1、x2The variable value is the fault strike value.
Thus, the cluster center calculation formula is as follows:
in the formula:
wherein c is a cluster center value, xiThe method comprises the steps of clustering sample values in a cluster, wherein n is the number of the samples in the cluster, the cluster is a group in a clustering result obtained after clustering analysis, and the clustering result is generally aggregated according to corresponding characteristics and is expressed as the cluster.
In step 130, clustering each dominant trend cluster again by taking the distance defined for the fault as a characteristic on the basis of fault trend characteristic clustering to obtain each fault zone developed in the well field, wherein the optimal grouping number of the fault distance characteristic is determined by a contour coefficient;
in step 140, a linear regression method is used to obtain the central line of each fault zone, and the extension index, the buffer radius and the expected fall attribute value of each fault zone are calculated to characterize the extension, the dispersion of the trend and the fall of the fault zone, and thereby predict the hidden fault in front of the mining working face.
In order to ensure the stability and reliability of the clustering result, the embodiment of the invention adopts a k-means + + method to select the initial clustering center, and adopts repeated clustering calculation and a method of obtaining an optimal result, namely, clustering calculation is carried out on the same group of data for multiple times, the clustering center of each time is recorded, and finally the optimal result obtained by comparison is adopted, wherein the index adopted by comparison is the profile coefficient.
And when a first clustering result clustered according to the trend characteristic value is obtained, evaluating the clustering analysis result of the fault distance characteristic by using the contour coefficient as an index, and performing multiple calculations to obtain the optimal number of fault zones of each group of fault sample data, namely each fault zone developed in the well field, from the clustering results of the clustering analysis performed on a plurality of groups of distance characteristic values.
Before calculating the extension index, the buffer radius and the fall expectation attribute values of each fault zone, the method further comprises the step of respectively defining the extension index, the buffer radius and the fall expectation of each fault zone, and specifically comprises the following steps: defining the elongation index as the exposed proportion of the fault zone in the mined range; defining the buffer radius as the swing radius of the construction belt around the construction midline; and defining the throw expectations as the expected values for the set of fault throws, as described in detail below.
According to the method provided by the embodiment of the invention, the fault is divided into three levels according to the influence degree of the fault on mining, the fault which has a certain influence on production is taken as valuable fault sample data, multiple clustering is carried out by using the trend characteristic and the distance characteristic in the subsequent analysis process, the attribute value of the fault zone is determined from the clustering result, namely the clustering result, and the prediction results of the extension, the dispersion of the trend and the fall characteristic of the fault zone are described, so that the prediction of the hidden fault in the mining area is realized, and the dynamic prediction of the fault in front of the mining working face is also realized by combining with the production dynamic recognition of the fault zone.
In an implementation scenario of the embodiment of the present invention, as shown in fig. 2, before clustering again on the distance defined for the fault in each dominant trend cluster, the method further includes a step of determining a reference line, where the process includes the following steps
In step 210, determining the datum line according to the determined mining area and the strike value of the fault zone of the mining area, including: in step 211, when the trend value of the construction line is within a first preset range, determining the reference line by combining the spatial coordinates of one end point of the mining area and the trend average value of all faults of each set of fault sample data; in step 212, when the constructed line trend value is in a second preset range, determining the datum line by combining the spatial coordinates of the other end point of the mining area and the trend average value of all faults of each set of fault sample data.
Referring to fig. 3, which is a schematic flow chart illustrating an implementation of the blind fault prediction method according to an embodiment of the present invention, first, data preprocessing is performed on the obtained exposure information, feature normalization is performed according to features of fault data of the exposure information, a level of each fault is determined according to a degree of influence of a fall on a coal seam thickness, and fault sample data used for clustering is further determined.
In the characteristic learning stage after the fault sample data is determined, the number of clustering groups (clusters) is determined by taking the contour coefficient as an evaluation index, clustering is carried out according to the trend characteristic value of each fault sample data to obtain a first clustering result, clustering analysis of the distance characteristic value is defined on the basis, including distance characteristic definition, second clustering result determination, clustering according to coordinates (positions) and fault zone linear regression are carried out to determine fault zone clustering results, attributes of fault zones are calculated to obtain at least extension indexes and possibly existing buffer radii, fall expectation and other attribute values, and finally prediction of the hidden fault position and other related attributes is realized according to the attribute values.
The fault band attribute includes at least an extension index of the fault band, the extension index being a probability that the fault band of a mining zone extends from a mined zone to an unextracted zone.
The method further comprises the steps of determining the dispersion of sample data of each fault included in each fault zone through the attribute of the fault zone by the dispersion, wherein the dispersion is the ratio of the sample standard deviation of the distance of a fault offset structure central line to the projection length of all faults included in the fault zone on the structure central line; and predicting whether the fault precision meets the requirement of guiding production according to the relation between the dispersion of the fault zone classification result and a preset value.
Referring to fig. 4, which is a schematic diagram of a reference line of an embodiment of the present invention, a rectangular area in fig. 4 is used as a mining area, when a trend value of a construction line is [0, 0.5 pi ], L1 in the drawing is used as a reference line, the line passes through a point (xmax, ymin), the point is a two-dimensional space coordinate of the mining area at an end point, and a slope is calculated from an average trend value of all faults in the cluster; when the trend value of the construction line is [0.5 pi, pi ], taking L2 in the graph as a reference line, wherein the line passes through a point (xmin, ymin), the point is a two-dimensional space coordinate of the mining area on the other end point, the slope is calculated from the average trend value of all faults of the cluster, and the distance characteristic value of each fault sample data can be obtained by combining the reference line with the coordinate of the fault sample data.
The dispersion (discrete degree) in the present invention is defined as the ratio of the sample standard deviation of the distance of the fault offset structure central line to the projection length of all faults in the cluster on the structure central line (the extension length of the structure line), and its calculation formula is as follows:
in the formula (4), ddFor dispersion, σ is the sample standard deviation of the fault-to-structure centerline distance, LiFor the projected length of the i-th slice on the construction line, LrTo construct a set of line segments of the line within the extraction envelope. In one implementation of the embodiment of the present invention, the mining area may be an irregular mining area, and the structural line may be divided into a plurality of line segments in the mining area, so that in this case, the intersection of the line segment set and the fault projection line segment set needs to be calculated.
The dispersion of the clusters obtained by clustering in the fault zone clustering result is calculated, the dispersion is large, the larger the horizontal swing distance deviating from the structural central line on the unit trend length is, the poorer the linear correlation degree is, and when the dispersion is larger than 0.2, the central line obtained by the fault regression can not effectively guide the coal mine production.
For fault zones with dispersion meeting the requirements, a linear regression equation is obtained by taking a node set of all fault lines in a cluster as sample points to serve as a structural centerline equation, a least square method is adopted to obtain an analytic expression of the linear equation, and coefficients are calculated as follows:
θ＝(XTWX)-1XTWY (5)
the elongation index is defined as the exposed proportion of the fault zone in the mined range, and the calculation formula is as follows:
wherein Lr is the projection length of the set of fault on the structure central line, L is the length of the fault structure straight line in the range of the excavation area, edReflecting the stability of the extension of the fault structure in the extraction area.
The damping radius is defined as the swing radius of the fault band about its structural centerline. According to the normal distribution characteristics, the probability of 95% (σ is the standard deviation of the distance characteristic value) of the distance within the range of [ μ -1.96 σ, μ +1.96 σ ], is calculated by the following formula (7): .
Wherein n is the number of the fault layers in the fault zone; diThe distance to the fault band centerline for the ith fault is shown. The drop height expectation is defined as the expectation value of the set of fault drop heights, is used for representing the drop height of the fault zone structure line, and is calculated by the following formula (8):
wherein n is the number of the fault layers in the fault zone; t is tiIndicating the throw of the ith fault.
The fault prediction mainly comprises the distance between a fault possible development position and a current working point, fault trend and fault zone trend. The characteristics of fault zones directly determine the direction and fall of faults. Assuming point a as the current operating point, the future 2210 lane will intersect the fracture zone centerline at point B, as shown in fig. 5. The distance between the point A and the point B is the fault position to be predicted.
The fault zone extension index represents the probability of an implosion fault being present in front of the working face. The buffer radius needs to be converted to a distance along the course of the roadway because the heading direction is not always perpendicular to the centerline of the fault zone. As shown in fig. 5, R represents the buffer radius and α represents the angle between the two lines. The deviation distance (d) is calculated as follows:
and the reliability of the prediction result is represented by adopting fault zone attributes, the extension index represents a probability theory of the existence of the predicted fault, and the buffer radius represents the possible offset distance of the predicted fault relative to the central line of the fault zone.
In a more specific implementation manner of the embodiment of the invention, as shown in fig. 6, blind fault prediction is performed by taking a certain coal mine 302 mining area as an example, the average coal thickness of the mining area is 5 meters, 14 fully mechanized mining working faces are designed 8202-8228, 4 working faces are currently mined 8212, 8214, 8216 and 8218, the working faces 8202 and 8220 are defined, and two crossroads 5210 and 2210 of the working face 8210 are being tunneled, which is specifically shown in fig. 6. In the mining process of the mining area, 86 fault lines are exposed, the fall is 0.5-8.0 m, the trend is 30-173 degrees, and the data of each fault line are detailed in a table 1.
TABLE 1 Fault data
Firstly, dividing all faults disclosed in a research area into 3 grades according to two characteristics of fall and extension length (obtained by combining fault coordinates which can be two-dimensional or three-dimensional coordinates according to actual parameters), wherein two thresholds of fall grading are 2.5 meters and 1 meter respectively. The results of the fractionation are shown in Table 2.
TABLE 2 results of fault classification
Secondly, identifying the trend of a fault structural zone;
(1) a plurality of 2-10 groups of clustering contour coefficients are respectively calculated, the values of the contour coefficients are used for indicating the aggregation degree of the clustering results, when the contour coefficients are higher, the clustering results are better, as shown in fig. 7, judgment is performed according to the values of the contour coefficients of each group of clustering, when the clustering is 3 groups, the contour coefficients reach the highest value of 0.641, and the contour coefficients under other grouping conditions are all lower than 0.641, so that 3 is taken as the optimal grouping number of the embodiment of the invention.
(2) According to the optimal grouping number in the step (1), fault samples are grouped into 3 groups for clustering analysis, the clustering centers of the first trend characteristic values, namely fault dominant trends, are respectively 61.3 degrees, 111.1 degrees and 161.1 degrees, wherein the clustering centers are mean values obtained by the quotient of the sum of the trend characteristic values of all fault samples in each group and the number of the fault samples in the group, the clustering result in the specific embodiment of the invention is shown in figure 8, red, blue and green lines in figure 8 represent grouping marks of the optimal grouping, and the lengths of fault samples in the figure can be calculated according to the coordinates.
Thirdly, determining the position of the fault structure zone
(1) After the trend clustering result is determined, the datum line of each group of trend clusters is calculated, and the distance between each fault and the corresponding datum line is obtained. The baseline parameters are shown in table 3.
TABLE 3 Baseline parameters
(2) Distance cluster analysis
And respectively calculating 2-11 groups of numerical clustering profile coefficients by taking each strike cluster fault sample as a unit, wherein the results are shown in fig. 9(a)/9(b)/9(c), and the number of corresponding strike cluster fault zones can be determined to be 4, 5 and 2 respectively.
And determining the fault zone of each trend cluster by combining the reference line, the number of trend cluster groups and the distance characteristic value of each trend cluster, wherein a fault zone clustering result diagram correspondingly obtained in fig. 9(a)/9(b)/9(c) is shown in fig. 10(a)/10(b)/10 (c).
(3) Fault zone centerline regression
Firstly, calculating the dispersion of all fault samples in each fault zone, as shown in table 4, if the dispersion is greater than 0.2, obtaining a fault zone central line linear equation by linear regression of each node of the fault line, and combining the fault data in tables 1 to 3 to obtain:
TABLE 4 results of the dispersion calculation
(4) Fault zone attribute calculation
The results of calculating the elongation index, the buffer radius, and the drop height of each fault zone expect 3 attribute values are shown in tables 5, 6, and 7.
TABLE 5 result of calculation of elongation index
TABLE 6 buffer radius calculation results
TABLE 7 drop expectation calculation results
Substituting the required tomographic data in combination with the above equation (6)The elongation index e can be obtaineddThe extension index may reflect the stability of the fault formation extension within the mining area;
the buffer radius can be calculated by the above equation (7).
The throw expectation is defined as the expected value of the set of fault throws, which is used to characterize the magnitude of the throw of the fault zone formation line, and can be calculated by the above equation (8).
And 4, step 4: predicting hidden faults
The predicted blind fault includes fault location and drop height, the location being measured in distance from the point of the mining operation, as shown in fig. 11, 5210 a roadway 404 m ahead in the heading direction may reveal a fault zone that is the 2 nd distance cluster of the 2 nd strike cluster. The relevant attributes can be found from tables 5, 6 and 7, and the extension index is 0.874, the buffer radius is 172.155, and the fall is 2.14 m, namely the probability of existence of the fault is 87.4%, the offset distance of the fault along the center line is in the range of 172.155 m, and the fault fall is 2.14 m, which is found by combining tables 5, 6 and 7.
Fig. 12 is a schematic structural diagram of an implicit fault prediction apparatus based on feature learning according to an embodiment of the present invention, where the apparatus may be implemented by software and/or hardware, is generally integrated in an intelligent terminal, and may be implemented by an implicit fault prediction method based on feature learning. As shown in the figure, the present embodiment may provide a hidden fault prediction apparatus based on feature learning based on the above embodiments, which mainly includes a data processing module 1110, a first clustering module 1120, a second clustering module 1130, and an analysis prediction module 1140.
The data processing module 1110 is configured to obtain fault attribute information according to the disclosure information of the mining area, where the fault attribute information at least includes fault position, fall, trend, and extension length attribute information, and process the fault attribute information according to a preset rule to obtain fault sample data;
the first clustering module 1120 is used for clustering fault sample data according to trend characteristics to obtain each dominant trend cluster of the well field fault development, and the optimal grouping number of the trend characteristics is determined by a contour coefficient;
the second clustering module 1130 is configured to cluster each dominant trend cluster again by using the distance defined for the fault as a feature on the basis of fault trend feature clustering to obtain each fault zone developed in the well field, where the optimal grouping number of the fault distance features is determined by a contour coefficient;
the analysis and prediction module 1140 is configured to obtain a central line of each fault zone by using a linear regression method for each fault zone obtained by the second clustering module, and calculate an extension index, a buffer radius, and a fall expectation attribute value of each fault zone to characterize extension, dispersion of trend, and fall characteristics of the fault zone, and predict the blind fault in front of the mining face by using the extension, dispersion, and fall characteristics.
In an implementation scenario of the embodiment of the present invention, the apparatus further includes:
a definition module, configured to respectively define an extension index, a buffer radius, and a drop expectation of each fault zone, including:
defining the extension index, wherein the extension index is the exposure proportion of the fault zone in the mined range;
defining the extension index, wherein the extension index is the swing radius of the fault belt in the fault belt structure middle line; and
defining the throw expectation, which is an expected value of a fault throw included in the fault zone.
In an implementation scenario of the embodiment of the present invention, the apparatus further includes:
the datum line determining module is used for determining the datum line according to the determined mining area and the trend value of the fault zone of the mining area; the datum line determining module comprises:
the first sub-module is used for determining the datum line by combining the space coordinate of one end point of the mining area and the trend average value of all faults of each set of fault sample data when the trend value of the construction line is in a first preset range;
and the second sub-module is used for determining the datum line by combining the space coordinate of the other end point of the mining area and the trend average value of all faults of each group of fault sample data when the trend value of the construction line is in a second preset range.
In an implementation scenario of the embodiment of the present invention, the analysis and prediction module is further configured to:
determining the dispersion of sample data of each fault included in each fault zone, wherein the dispersion is the ratio of the sample standard deviation of the distance of the fault offset construction central line to the projection length of all faults included in the fault zone on the construction central line;
and predicting whether the fault precision meets the requirement of guiding production according to the relation between the dispersion of the fault zone classification result and a preset value.
In an implementation scenario of the embodiment of the present invention, the data processing module 1110 includes:
taking the ratio of fault fall to the average coal thickness of the mining working face and the extension length of the fault as characterization indexes of the fault data;
processing the characterization indexes of the fault data according to a piecewise linear normalization processing device based on a first preset threshold and a second preset threshold to obtain fault development scale weight;
and clustering the fault data by taking the fault development scale weight as a clustering feature, and taking the fault which has a certain influence on the production in a clustering result as fault sample data of next-step clustering analysis.
The hidden fault prediction device based on feature learning provided in the above embodiment can execute the hidden fault prediction method based on feature learning provided in any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method.
The technical carrier involved in payment in the embodiments of the present specification may include Near Field Communication (NFC), WIFI, 3G/4G/5G, POS machine card swiping technology, two-dimensional code scanning technology, barcode scanning technology, bluetooth, infrared, Short Message Service (SMS), Multimedia Message (MMS), and the like, for example.
The biometric features related to biometric identification in the embodiments of the present specification may include, for example, eye features, voice prints, fingerprints, palm prints, heart beats, pulse, chromosomes, DNA, human teeth bites, and the like. Wherein the eye pattern may include biological features of the iris, sclera, etc.
It should be noted that the method of one or more embodiments of the present disclosure may be performed by a single device, such as a computer or server. The method of the embodiment can also be applied to a distributed scene and completed by the mutual cooperation of a plurality of devices. In such a distributed scenario, one of the plurality of devices may perform only one or more steps of the method according to one or more embodiments of the present disclosure, and the plurality of devices may interact with each other to complete the feature learning-based blind fault prediction method.
The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
For convenience of description, the above devices are described as being divided into various modules by functions, and are described separately. Of course, the functionality of the modules may be implemented in the same one or more software and/or hardware implementations in implementing one or more embodiments of the present description.
The apparatus of the foregoing embodiment is used to implement the corresponding method in the foregoing embodiment, and has the beneficial effects of the corresponding method embodiment, which are not described herein again.
Fig. 13 is a schematic diagram illustrating a more specific hardware structure of an electronic device according to this embodiment, where the electronic device may include: a processor 1010, a memory 1020, an input/output interface 1030, a communication interface 1040, and a bus 1050. Wherein the processor 1010, memory 1020, input/output interface 1030, and communication interface 1040 are communicatively coupled to each other within the device via bus 1050.
The processor 1010 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an Application Specific Integrated Circuit (ASIC), or one or more Integrated circuits, and is configured to execute related programs to implement the technical solutions provided in the embodiments of the present disclosure.
The Memory 1020 may be implemented in the form of a ROM (Read Only Memory), a RAM (Random access Memory), a static storage device, a dynamic storage device, or the like. The memory 1020 may store an operating system and other application programs, and when the technical solution provided by the embodiments of the present specification is implemented by software or firmware, the relevant program codes are stored in the memory 1020 and called by the processor 1010 to execute the feature learning-based blind fault prediction method.
The input/output interface 1030 is used for connecting an input/output module to input and output information. The i/o module may be configured as a component in a device (not shown) or may be external to the device to provide a corresponding function. The input devices may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output devices may include a display, a speaker, a vibrator, an indicator light, etc.
The communication interface 1040 is used for connecting a communication module (not shown in the drawings) to implement communication interaction between the present apparatus and other apparatuses. The communication module can realize communication in a wired mode (such as USB, network cable and the like) and also can realize communication in a wireless mode (such as mobile network, WIFI, Bluetooth and the like).
Bus 1050 includes a path that transfers information between various components of the device, such as processor 1010, memory 1020, input/output interface 1030, and communication interface 1040.
It should be noted that although the above-mentioned device only shows the processor 1010, the memory 1020, the input/output interface 1030, the communication interface 1040 and the bus 1050, in a specific implementation, the device may also include other components necessary for normal operation. In addition, those skilled in the art will appreciate that the above-described apparatus may also include only those components necessary to implement the embodiments of the present description, and not necessarily all of the components shown in the figures.
Computer-readable media of the present embodiments, including both non-transitory and non-transitory, removable and non-removable media, may implement information storage by any method or technology. The information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), other types of Random Access Memory (RAM), Read Only Memory (ROM), Electrically Erasable Programmable Read Only Memory (EEPROM), flash memory or other memory technology, compact disc read only memory (CD-ROM), Digital Versatile Discs (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device.
Those of ordinary skill in the art will understand that: the discussion of any embodiment above is meant to be exemplary only, and is not intended to intimate that the scope of the disclosure, including the claims, is limited to these examples; within the spirit of the present disclosure, features from the above embodiments or from different embodiments may also be combined, steps may be implemented in any order, and there are many other variations of different aspects of one or more embodiments of the present description as described above, which are not provided in detail for the sake of brevity.
In addition, well-known power/ground connections to Integrated Circuit (IC) chips and other components may or may not be shown in the provided figures, for simplicity of illustration and discussion, and so as not to obscure one or more embodiments of the disclosure. Furthermore, devices may be shown in block diagram form in order to avoid obscuring the understanding of one or more embodiments of the present description, and this also takes into account the fact that specifics with respect to implementation of such block diagram devices are highly dependent upon the platform within which the one or more embodiments of the present description are to be implemented (i.e., specifics should be well within purview of one skilled in the art). Where specific details (e.g., circuits) are set forth in order to describe example embodiments of the disclosure, it should be apparent to one skilled in the art that one or more embodiments of the disclosure can be practiced without, or with variation of, these specific details. Accordingly, the description is to be regarded as illustrative instead of restrictive.
While the present disclosure has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those of ordinary skill in the art in light of the foregoing description. For example, other memory architectures (e.g., dynamic ram (dram)) may use the discussed embodiments.
It is intended that the one or more embodiments of the present specification embrace all such alternatives, modifications and variations as fall within the broad scope of the appended claims. Therefore, any omissions, modifications, substitutions, improvements, and the like that may be made without departing from the spirit and principles of one or more embodiments of the present disclosure are intended to be included within the scope of the present disclosure.
Claims (10)
1. A method for feature learning based blind fault prediction, the method comprising:
acquiring fault attribute information according to the exposure information of the mining area, wherein the fault attribute information at least comprises fault position, fall, trend and extension length attribute information, and processing the fault attribute information according to a preset rule to acquire fault sample data;
clustering the fault sample data by using trend characteristics to obtain each dominant trend cluster of the well field fault development, wherein the optimal grouping number of the trend characteristics is determined by a contour coefficient;
clustering each dominant trend cluster again by taking the distance defined for the fault as a characteristic on the basis of fault trend characteristic clustering to obtain each fault zone developed in a well field, wherein the optimal grouping number of the fault distance characteristic is determined by a contour coefficient;
and obtaining the central line of each fault zone by adopting a linear regression method, calculating the extension index, the buffer radius and the expected fall attribute value of each fault zone to characterize the extension, the dispersion of the trend and the fall of the fault zone, and predicting the hidden fault in front of the mining working face.
2. The method of claim 1, wherein prior to calculating the elongation index, buffer radius, and fall expectation attribute values for each fault zone, the method further comprises:
defining the extension index, the buffer radius and the drop height expectation of each fault zone respectively, wherein the definition comprises the following steps:
defining the extension index, wherein the extension index is the exposure proportion of the fault zone in the mined range;
defining the extension index, wherein the extension index is the swing radius of the fault belt in the fault belt structure middle line; and
defining the throw expectation, which is an expected value of a fault throw included in the fault zone.
3. The method of claim 1, wherein before clustering each dominant trend cluster again characterized by a fault-defined distance, the method further comprises:
determining the datum line according to the determined mining area and the trend value of the fault zone of the mining area, wherein the datum line comprises the following steps:
when the trend value of the construction line is in a first preset range, determining the datum line by combining the space coordinate of one end point of the mining area and the trend average value of all faults of each set of fault sample data;
and when the trend value of the construction line is in a second preset range, determining the datum line by combining the spatial coordinate of the other end point of the mining area and the trend average value of all faults of each set of fault sample data.
4. The method of claim 1, further comprising:
determining the dispersion of sample data of each fault included in each fault zone, wherein the dispersion is the ratio of the sample standard deviation of the distance of the fault offset construction central line to the projection length of all faults included in the fault zone on the construction central line;
and predicting whether the fault precision meets the requirement of guiding production according to the relation between the dispersion of the fault zone classification result and a preset value.
5. The method of claim 1, wherein the processing the fault data according to a preset rule to obtain fault sample data comprises:
taking the ratio of fault fall to the average coal thickness of the mining working face and the extension length of the fault as characterization indexes of the fault data;
processing the characterization indexes of the fault data according to a piecewise linear normalization processing method based on a first preset threshold and a second preset threshold to obtain fault development scale weight;
and clustering the fault data by taking the fault development scale weight as a clustering feature, and taking the fault which has a certain influence on the production in a clustering result as fault sample data of next-step clustering analysis.
6. An apparatus for feature learning based blind fault prediction, the apparatus comprising:
the data processing module is used for obtaining fault attribute information according to the exposure information of the mining area, wherein the fault attribute information at least comprises fault position, fall, trend and extension length attribute information, and processing the fault attribute information according to a preset rule to obtain fault sample data;
the first clustering module is used for clustering fault sample data according to trend characteristics to obtain each dominant trend cluster of the well field fault development, and the optimal grouping number of the trend characteristics is determined by a contour coefficient;
the second clustering module is used for clustering each dominant trend cluster again by taking the distance defined for the fault as a characteristic on the basis of fault trend characteristic clustering to obtain each fault zone developed in the well field, wherein the optimal grouping number of the fault distance characteristic is determined by a contour coefficient;
and the analysis and prediction module is used for obtaining the central line of each fault zone by adopting a linear regression method for each fault zone obtained by the second clustering module, calculating the extension index, the buffer radius and the expected fall attribute value of each fault zone, describing the extension, the dispersion of the trend and the fall characteristic of the fault zone, and predicting the hidden fault in front of the mining working face.
7. The apparatus of claim 6, further comprising:
a definition module, configured to respectively define an extension index, a buffer radius, and a drop expectation of each fault zone, including:
defining the extension index, wherein the extension index is the exposure proportion of the fault zone in the mined range;
defining the extension index, wherein the extension index is the swing radius of the fault belt in the fault belt structure middle line; and
defining the throw expectation, which is an expected value of a fault throw included in the fault zone.
8. The apparatus of claim 6, further comprising:
the datum line determining module is used for determining the datum line according to the determined mining area and the trend value of the fault zone of the mining area; the datum line determining module comprises:
the first sub-module is used for determining the datum line by combining the space coordinate of one end point of the mining area and the trend average value of all faults of each set of fault sample data when the trend value of the construction line is in a first preset range;
and the second sub-module is used for determining the datum line by combining the space coordinate of the other end point of the mining area and the trend average value of all faults of each group of fault sample data when the trend value of the construction line is in a second preset range.
9. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method according to any of claims 1 to 5 when executing the program.
10. A non-transitory computer-readable storage medium storing computer instructions for causing a computer to perform the method of any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010206984.4A CN111582536B (en) | 2020-03-23 | 2020-03-23 | Feature learning-based hidden fault prediction method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010206984.4A CN111582536B (en) | 2020-03-23 | 2020-03-23 | Feature learning-based hidden fault prediction method, device, equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111582536A true CN111582536A (en) | 2020-08-25 |
| CN111582536B CN111582536B (en) | 2024-01-30 |
Family
ID=72120526
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010206984.4A Active CN111582536B (en) | 2020-03-23 | 2020-03-23 | Feature learning-based hidden fault prediction method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111582536B (en) |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE719105C (en) * | 1929-04-13 | 1942-03-28 | Mitteldeutsche Stahlwerke Ag | Plant for the extraction and conveyance of overburden in opencast mines |
| CA2575066A1 (en) * | 2006-01-24 | 2007-07-24 | The University Of North Carolina At Chapel Hill | Systems and methods for detecting an image of an object by use of an x-ray beam having a polychromatic distribution |
| AU2009300343A1 (en) * | 2008-10-02 | 2010-04-08 | Certusview Technologies, Llc | Methods and apparatus for analyzing locate and marking operations with respect to historical information |
| CN102156779A (en) * | 2011-04-13 | 2011-08-17 | 北京石油化工学院 | Subsurface flow simulating and predictive analysis method |
| AU2012200758A1 (en) * | 2005-03-22 | 2012-03-01 | Ticketmaster | Apparatus and methods for providing queue messaging over a network |
| CN103150335A (en) * | 2013-01-25 | 2013-06-12 | 河南理工大学 | Co-clustering-based coal mine public sentiment monitoring system |
| CN106885772A (en) * | 2015-12-15 | 2017-06-23 | 核工业北京地质研究院 | The ETM remote sensing hydroxyls altered mineral that a kind of geologic knowledge is participated in extracts improved method |
| CN107169616A (en) * | 2017-07-21 | 2017-09-15 | 西安科技大学 | Mine unworked country constructs the relative entropy Forecasting Methodology of relative component degree |
| US20170356955A1 (en) * | 2016-06-14 | 2017-12-14 | Freescale Semiconductor, Inc. | Kernel based cluster fault analysis |
| US20190107643A1 (en) * | 2017-10-11 | 2019-04-11 | Beyond Limits, Inc. | Static engine and neural network for a cognitive reservoir system |
-
2020
- 2020-03-23 CN CN202010206984.4A patent/CN111582536B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE719105C (en) * | 1929-04-13 | 1942-03-28 | Mitteldeutsche Stahlwerke Ag | Plant for the extraction and conveyance of overburden in opencast mines |
| AU2012200758A1 (en) * | 2005-03-22 | 2012-03-01 | Ticketmaster | Apparatus and methods for providing queue messaging over a network |
| CA2575066A1 (en) * | 2006-01-24 | 2007-07-24 | The University Of North Carolina At Chapel Hill | Systems and methods for detecting an image of an object by use of an x-ray beam having a polychromatic distribution |
| AU2009300343A1 (en) * | 2008-10-02 | 2010-04-08 | Certusview Technologies, Llc | Methods and apparatus for analyzing locate and marking operations with respect to historical information |
| CN102156779A (en) * | 2011-04-13 | 2011-08-17 | 北京石油化工学院 | Subsurface flow simulating and predictive analysis method |
| CN103150335A (en) * | 2013-01-25 | 2013-06-12 | 河南理工大学 | Co-clustering-based coal mine public sentiment monitoring system |
| CN106885772A (en) * | 2015-12-15 | 2017-06-23 | 核工业北京地质研究院 | The ETM remote sensing hydroxyls altered mineral that a kind of geologic knowledge is participated in extracts improved method |
| US20170356955A1 (en) * | 2016-06-14 | 2017-12-14 | Freescale Semiconductor, Inc. | Kernel based cluster fault analysis |
| CN107169616A (en) * | 2017-07-21 | 2017-09-15 | 西安科技大学 | Mine unworked country constructs the relative entropy Forecasting Methodology of relative component degree |
| US20190107643A1 (en) * | 2017-10-11 | 2019-04-11 | Beyond Limits, Inc. | Static engine and neural network for a cognitive reservoir system |
Non-Patent Citations (8)
| Title |
|---|
| QIANG WU ET AL: ""Prediction of concealed faults in front of a coalface using feature learning"", 《BULLETIN OF ENGINEERING GEOLOGY AND THE ENVIRONMENT》, pages 4191 * |
| 刘德民等: ""团柏煤矿下组煤开采矿井充水条件分析"", 《煤矿安全》, vol. 46, no. 1, pages 177 - 179 * |
| 刘敬寿;丁文龙;肖子亢;戴俊生;: "储层裂缝综合表征与预测研究进展", 《地球物理学进展》, vol. 34, no. 6, pages 2283 - 2300 * |
| 刘栋等: ""用沙坝断层滑移型微震事件群活动特征分析"", 《安全与环境学报》, vol. 18, no. 3, pages 1036 - 1040 * |
| 孙振宇等: ""基于SVM算法的地震小断层自动识别"", 《煤炭学报》, vol. 42, no. 11, pages 2945 - 2952 * |
| 来骥;盛红雷;: "基于聚类分析的复杂网络链路预测性能研究", 《计算技术与自动化》, vol. 38, no. 4, pages 144 - 150 * |
| 胡幸平;崔效锋;: "华北地区中部地震精定位与构造应力场研究", 《震灾防御技术》, vol. 8, no. 4, pages 351 - 360 * |
| 郝治朝: ""煤矿隐伏断层定量预测方法与应用研究"", 《中国博士学位论文全文数据库》, pages 1 - 117 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111582536B (en) | 2024-01-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102236046B1 (en) | Face detection training method, device and electronic device | |
| US20190050981A1 (en) | System and method for automatically detecting a target object from a 3d image | |
| Li et al. | Coal gangue detection and recognition algorithm based on deformable convolution YOLOv3 | |
| CN111523678B (en) | Service processing method, device, equipment and storage medium | |
| CN115668308A (en) | Automatic detection of terrain-based wellsites and their perimeters | |
| CN106558096A (en) | The modeling method of the model building device, program and cage guide of three dimensional point cloud | |
| EP3938814B1 (en) | Automated facies classification from well logs | |
| CN115240149A (en) | Three-dimensional point cloud detection and identification method and device, electronic equipment and storage medium | |
| Gómez et al. | Determining the accuracy in image supervised classification problems | |
| Crawford et al. | Using landslide-inventory mapping for a combined bagged-trees and logistic-regression approach to determining landslide susceptibility in eastern Kentucky, USA | |
| CN115830010B (en) | Method and device for detecting crossing distance of transmission wires and computer equipment | |
| Zhao et al. | Evaluation of convolution operation based on the interpretation of deep learning on 3-d point cloud | |
| Wang et al. | Automatic identification and location of tunnel lining cracks | |
| CN109492775A (en) | A kind of detection method of geologic structure interpretation, detection device and readable storage medium storing program for executing | |
| CN111582536A (en) | Hidden fault prediction method, device, equipment and medium based on feature learning | |
| CN111462490B (en) | Road network visualization method and device based on multistage subregion division | |
| US20230093528A1 (en) | Interpretation method for neural network model, electronic device and storage medium | |
| CN111582647A (en) | User data processing method and device and electronic equipment | |
| CN111598384A (en) | Method, device, equipment and medium for determining variable weight interval threshold based on clustering coupling | |
| CN113792961A (en) | Economic responsibility auditing decision method and equipment based on big audit data | |
| US11755948B2 (en) | Attribution and generation of saliency visualizations for machine-learning models | |
| CN114548307A (en) | Classification model training method and device, and classification method and device | |
| CN114360735A (en) | Infectious disease time-space aggregation detection and analysis method and system and electronic equipment | |
| RU2530324C2 (en) | Method for determining position of marker depth coordinates when constructing geological model of deposit | |
| CN116630752B (en) | Construction site target object identification method and device based on AI algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |